An Enhanced ML based Approach for Stream Selection of Higher Secondary Education in India: CareerX

Abstract

In the Indian education system, choosing right stream for higher secondary is critical for shaping the future careers and job prospects of the students. It can be overwhelming, as many young students have not yet developed clear interests. Using machine learning approaches, we have created a model to increase the efficiency of this selection. Data, such as academic scores, IQ test results, and personality tests labelled with the students' selected courses, were gathered from multiple sites in Gujarat, India. IQ test results, SSC and SSC grade standard scores were initially used to train the model. Subsequently, we incorporated personality traits into the model to examine their impact on stream selection. Lastly, we only used personality factors to train the models. Accuracy, precision, recall, and F1-score were among the criteria used to assess these model’s performance. Our findings demonstrate that the models effectively predict appropriate streams for students, providing a standardized and data-driven approach to streamline the decision-making process. Interestingly, the findings demonstrate how personality traits have a big impact on stream choice. Based on their academic achievement, IQ, and personality traits, this framework has the potential to help students make more informed and individualized educational decisions.